Efficacy and Complications of Percutaneous Pigtail Catheters for Thoracostomy in Pediatric Patients*

Transcription

1 Efficacy and Complications of Percutaneous Pigtail Catheters for Thoracostomy in Pediatric Patients* Joan S. Roberts, MD; Susan L. Bratton, MD; and Thomas V. Brogan, MD Objective: To describe the efficacy of percutaneous pigtail catheters in evacuating pleural air or fluid in pediatric patients. Design: A case series of children with percutaneous pigtail catheters placed in the pediatric ICU between January 1996 and August Setting: U rhan pediatric teaching hospital in Seattle, W A. Methods: A retrospective chart review. Results: Ninety-one children required 133 chest catheters. Most patients were infants with congenital heart disease (80%). One hundred thirteen of the catheters (85%) were placed for pleural effusion, with 20 tubes (15%) placed fm pneumothorax. Efficacy of drainage of pleural fluid was significantly greater in serous (96%) and chylous (100%) effusions compared with empyema (0%) or hemothorax (81 %). Evacuation of pneumothorax was achieved by a pigtail catheter in 75% of patients. Resolution of pleural air or pneumothorax was significantly greater in patients < 10 kg compared with larger children. Complications due to placement of the pigtail catheters included hemothorax (n=3, 2%), pneumothorax (n=3, 2%), and hepatic perforation (n= 1, 1 %). There were also complications arising from the use of the catheters, including failure to drain, dislodgment, kinking, loss of liquid ventilation fluid, empyema, and disconnection in 27 of 133 catheters (20%). Significantly more complications dming catheter use occurred in patients < 5 kg than in larger children. Conclusions: Percutaneous pigtail catheters are highly effective in drainage of pleural serous and chylous effusions, somewhat less efficacious in drainage of hemothorax or pneumothorax, and least efficacious in drainage of empyema. Infants and smaller children had higher rates of resolution of pleural air and fluid f1 om placement of a pigtail catheter than larger children. Complications from catheter placement were uncommon (5%) hut serious, whereas complications associated with continued use of the catheters were more common (20%) hut less grave. Strict attention to anatomic landmarks and close monitoring may reduce the number of complications. (CHEST 1998; 114: ) Key words: chest tube; pediatric; pleural effusion; pneumothorax; tube thoracostomy Abbreviations: ECMO=extracorporeal membranous oxygenation; PT= prothrombin time; PTI=pmtial thromboplastin time Thoracostomy tubes are a mainstay of treatment for removing fluid or air from the pleural space. Placement of a chest tube is, however, an invasive procedure with potential morbidity. Complications include hemothorax, perforation of intra-abdominal or intrathoracic organs, diaphragmatic laceration, empyema, pulmonary edema, and Horner's syn- *From the Department of Anesthesiology, University of Washington School of Medicine, and Department of Anesthesia and Critical Care, Children's Hospital and Medical Center, Seattle, WA. Manuscript received F ebruary 12, 1998; revision accepted April 4, Correspondence to: Joan S. Roberts, MD, Children's Hospital and ihedical Center, PO Box 5371, Seattle, WA 98105, jreno@u. washington. edu drome. 1-3 In an effort to reduce these complications, Fuhrman et al 4 and subsequently Lawless et al 5 described the use of percutaneous pigtail catheters in place of traditional large-bore tubes for thoracostomy and pleural drainage. The Seldinger needleguide wire method of placement and smaller, more flexible catheters avoid the force required to place a large-bore chest tube by the dissection or trocar methods. Given the potential morbidity of traditional chest tube insertion, use of the pigtail catheter may be desirable. The purpose of this study is to determine the efficacy of pigtail thoracostomy catheters in a large sample of the pediatric population and to 1116 Clinical Investigations

2 investigate the nature and frequency of complications associated with their use. MATERIALS AND METHODS The charts of all pediatric patients at Children's Hospital am.l Medical Center, Seattle, WA, whose discharge diagnosis from the ICU included pneumothorax or pleural effusion were reviewej. All patients who unclerwent percutaneous pigtail thoracostomy tube placement as the initial therapy for pneumothorax or pleural effusion were induded. Patients treated with surgical c;hest tubes or pigtail catheters placej outside the ICU were excluded. Patient data fi mn Janumy 1996 through August 1997 were included. Data collected included demographic information, indication for thoracostomy tube plac;ement, patient ventilation and coagulation status, size and site of the chest catheter, sedation given during placement, level of training and specialty of physic;ian performing thoracostomy tube placement, chest c;atheter life, resolution of the effusion or pneumothorax, and complkations of placement or catheter use. Type of effusion was determined by clinical obsetvation at the time of pl ac;ement as serous, chylous, sanguinous, or pu!lllent. Prothrombin time (PT) and partial thromboplastin time (PTr) were recorded as measures of coagulation, with normal values for our laboratmy of PT of 11.3 to 17 s and PTI of 24 to 50 s. Autopsy findings were reviewed when available. Percutaneous pigtail catheters (Cook Ctitical Care; Cook lnc;orporated; Bloomington, IN) were all single-lumen polyurethane coiled catheters, 7 to 8.5F, used in c;onjunction with a wire and dilator, c;onnec;ted to a negative-pressure drainage syste m. The catheters were inserted using the modified Seldinger technique, with insertion of the needle and syringe over a rib, with gentle aspiration of a syringe to locate either fluid or air in the pleural space. A J-tipped wire was then inserted and the needle removed. A dilator and scalpel were used to enlarge the insertion site, and the catheter was then inserted ove r the wire. Finally the wire was removed and the catheter was attac;hed to a drain (Water-Seal Chest Drain; Atrium Medical Corporation; Hudson, NH). Bec;ause of difficulties securing the catheter to the chest, we used a modification of the c;onnection between the catheter and drainage tubing consisting of a length of extension tubing with a roller damp to avoid excessive torque and tension on the child's chest wall (Fig 1). Resolution of pleural fluid or air collection was defin ed as improvement in the effusion or pneumothorax clinically or by radiographic findings and that no other interventions were required. If the effi1sion or pneumothorax reaccumulated after the tube had been electively removed and an additional catheter was placed, the first episode of effusion or pneumothorax was recorded as resolved for study purposes. Categorical data were analyzed by the x 2, Fisher's Exact Test, and x 2 for trend tests. Continuous independent data were analyzed b y the Mann-Whitney U tes t. Significance was defined as p < A commercial software package (SPSS of Windows; SPSS Inc; Chicago, IL) was used for the data analysis. RESULTS Ninety-one children required 133 percutaneous pigtail catheters. The demographic data are shown in Table l. Most patients were infants and children following surge1y for congenital heart disease. Mechanical ventilation was used for most patients, 61 FIGURE l. Anesthesia extension tubing connected to the pigtail c;atheter allows less tension on the chest tube to decrease kinking of the catheter. The catheter is sutured at the skin. A d ear dressing is plac;ed over the c;atheter and the anesthesia tubing is also taped to the skin. (67% ). The patients required a moderate level of ventilatory support with a median fraction of inspired oxygen of 0.5 and a median positive endexpiratory pressure of 5 em H 2 0. Twenty-six of 91 children had a prolonged PT or PTT (29%) at the time of catheter insertion, 4 children (5%) were supp01ied with extracorporeal membranous oxygenation (ECMO), and the mortality rate was 14% (12 patients). The median length of stay in the ICU was 14 days, and median length of stay in the hospital was 23 days. Sixty-three patients had a single catheter Table!-Demographic Features in 91 Patients* Median or No. (Range or %) Age, yr 0.7 (0-18) Weight, kg 7 (1.8-66) PEEP, em H 2 0 t 5 (3-20) F10 2, o/ot 50 (21-100) Female, No. (%) 42 (46) Male, No. (%) 49 (54) Diagnosis, No. (%) Congenital heart disease 73 (80) Pneumonia 5 (6) ARDS 6 (7) Other 7 (7) Mechanical ventilation, No. (%) 61 (67) ECMO, No. (%) 4 (5) Coagu lopathy, No. (%) 39 (29) SU!vival, 0. (%) 78 (86) *PEEP= positive end-expiratmy pressure; FI0 2 = fraction of' inspired oxygen. tventilated patients. CHEST/114/4/0CTOBER,

3 placed, 19 patients had bilateral catheters, and 9 patients required multiple catheters over prolonged hospitalizations. Table 2 shows the characteristics of placement of thoracostomy tubes in our ICU. The most common indication for thoracostomy drainage was pleural effusion, with 113 of 133 (85%) tubes placed to relieve an effusion. Chest tube placement was facilitated by the use of both topical anesthetic (83%) and systemic opioids (72%), as well as neuromuscular blockade (32%). Sedative medications, including benzodiazepines, ketamine, or propofol were given in 67% of cases. Pediatric intensive care fellows performed most percutaneous thoracostomy tube placements (91%) in our ICU. Composition of pleural fluid was most commonly serous (64%), followed by chyle (18%), blood (14%), and empyema (4%) (Table 3). In all five patients with empyema, Gram-positive organisms were identified. Comparison of resolution of the various types of effusion showed significantly higher efficacy in serous (96%) and chylous (100%) effusions compared with both hemothorax (81 %) or empyema (0%). Adequate drainage was significantly related to the patient's size, with a resolution rate of 98% among patients <5 kg, 93% for patients 5 to 10 kg, and 75% for patients > 10 kg. Fifteen percent of catheters were placed for pneumothoraces. Resolution of pneumothorax occurred in 15 of 20 catheters placed (75%). There were nine patients <10 kg who had a catheter placed for a pneumothorax, with evacuation of air in all nine. Significantly, more catheters failed to resolve the pneumothorax in patients who Table 2-Characteristics of 133 Catheter Placements No. (%) Indication Pleural effus ion 113 (85) Pneumothorax 20 (15) Operator Cardiothoracic surgery 7 (5) Pediatrics 121 (91) Radiology (1) General surge1y 4 (3) Medications given for procedure Topical anesthetic 110 (83) Sedation 89 (67) Opiate 96 (72) NMBD* 43 (32) Catheter size 7.0F 74 (55) 8.5F 53 (40) Not recorded 6 (5) Insertion side Left side of chest 45 (33) Right side of chest 88 (66) * N M BD = neuromuscular blockade. Table 3-0utcome of 133 Thoracostomy Tube Placements Duration of placemeut. d Indication Effusion Serous Chylous Empyema Hemothorax Pneumothorax Resolution Eflusion Serous Chylous Empyema Hemothorax Pneumothorax Resolution by weight, kg > 10 No *Compared with serous and chylous effusions. tx 2 for trend test. Median 3 No No. Resolved (Range) (0-69) (%) (54) (15) (4) (ll) (1.5 ) (96) (100) (0)* (81)* (75)* (%) (98) (93) (75)t weighted > 10 kg, 4 of 11 (36%). There was no significant difference in resolution of pneumothorax or effusion related to the size of the catheter. Among patients requiring a replacement thoracostomy tube, both Argyle and pigtail catheters were used. The placement site for the pigtail catheter was the same regardless of the indication for thoracostomy tube. Complications are shown in Table 4, with 33 of 133 (25%) catheters resulting in some type of complication. Complications occurred significantly more frequently in infants <5 kg, 17 of 47 (36%), compared with toddlers 5 to 10 kg, 4 of 46 (9%), or larger children, 11 of 40 (28%). Rare but major com- Table 4-Complications of 133 Pigtail Catheters Complications at placement Hemothorax Hepatic perforation Pneumothorax Complications of use Failure to drain Compression by chest wall Disconnection of tubing Accidental dislodgment Kink in catheter Loss of liquid ventilation fluid Empyema No l (%) (3) (1) (3) (ll) (l) (l ) (4) (2) (l ) (1) 1118 Clinical Investigations

4 plications included cannulation of a hepatic vessel (n= 1, 1 %) (Fig 2), hemothorax (n=3, 2%), pneumothorax (n=3, 2%), and empyema (n=1, 1 %). The patient with hepatic injury required a catheterization procedure with embolic coiling of the catheter tract. Of the three patients with hemothorax, one required reintubation and emergency transfusion, another required placement with an Argyle chest tube, and the third patient ultimately died after an additional pigtail catheter had been placed with cessation of bleeding. The incidence of major complications was not related to size. No cases of hemothorax occurred in patients who were coagulopathic or receiving ECMO. Other complications included failure to drain the effusion or air requiring repositioning or replacement of the catheter (n = 15, 11% ), dislodgment (n=6, 5%), and kinks or disconnection of the tubing (n = 3, 2%). One patient with progressive ARDS treated with partial liquid ventilation developed a pneumothorax, with loss of perflubron through the pigtail catheter (Fig 3). This required substantial replacement of perflubron to maintain adequate filling. There were no fatalities directly related to thoracostomy tubes. Nine of the 12 patients who died had an autopsy performed, none of which demonstrated gross or microscopic injury attributable to the pigtail catheter. FIGURE 2. Chest radiograph d emonstrating a pigtail catheter traversing the liver and enteling the right atlium via a hepatic vessel. FIGURE 3. Chest radiograph of a child receiving liquid ventilation with perflubron leaking out the pigtail catheter; arrow indicates site of catheter. CONCLUSIONS The use of thoracostomy tubes for draining pleural fluid or air is an important therapeutic measure that ideally provides effective drainage in a timely manner without complications from the procedure. Traditional large-bore chest tubes, placed by either blunt dissection or by trocar assistance, may have significant morbidity associated with the force required to breech the chest wall and the stiffness of the chest tube itself. Chest tube placement in neonates is particularly difficult, given their pliant chest wall and the close proximity of vital structures. Development of a polyurethane pigtail catheter by Fuhrman et al 4 provided a potentially less traumatic alternative to the traditional method. In our experience, the catheters are simple to place in critically ill patients. We found that pigtail catheters were very effective in draining serous and chylous effusions, but had a substantial failure rate when draining blood or air, and no resolution in cases of empyema. Fuhrman et al 4 reported that 4 of 12 patients required further drainage procedures after initial pigtail catheter placement. Two patients had bronchopulmonary fistulas, one had a chylous effusion, and one had accidental catheter dislodgment. In contrast to our study, Fuhrman et al 4 reported resolution of empyema in two of two cases compared with failure in all five of our patients. When treatment of empyema requires drainage, we recommend initial placement of a large-bore chest tube for patients with empyema. Ramnath et al 6 CHEST / 114 / 4 / 0CTOBER,

5 recently reported the utility of sonographic evaluation of parapneumonic effusions to decide treatment options. This retrospective study suggests that patients with organized effusions had shorter hospital stays when surgically treated, whereas free-flowing effusions did not benefit from pleural drainage. Regardless of pleural drainage, a diagnostic thoracentesis for Gram's stain and culture is indicated in patients with significant respiratory distress. 7 Pleural air was effectively drained by the pigtail catheters in 75% of our patients. Lawless et al 5 reported similar results in a series of 16 patients with 18 catheters placed for pleural air or pneumomediastinum. There were only two failures, yielding an 88% resolution rate. The air evacuation rates from both our series and that of Lawless et al 5 are greater than previously reported with conventional chest tube drainage in neonates with pneumothoraces. Allen et al 8 reported that 44% of initial attempts to relieve neonatal pneumothorax were unsuccessful. In the four patients <5 kg with pneumothoraces in our study, all had resolution with placement of the pigtail catheter. Compmison of small-caliber chest tubes and standard chest tubes in adults has shown that smaller tubes are more likely to malfunction, and that efficacy of standard chest tubes for pneumothoraces in adults is about 85%. 9 Unlike conventional chest tubes, pigtail catheters are easily compressed. Children > 10 kg had a significantly higher failure rate compared with smaller children, which may be due to relatively thicker chest walls. The rate of major complications, including hemothorax, pneumothorax, and liver perforation, \Vas low (5%). This is a dramatic improvement compared with prior reports of lung perforation in neonates after chest tube placement. Moessinger et ap 0 found that 25% of autopsy specimens among neonates requiring chest tubes for pneumothorax had perforation of the lung parenchyma. The incidence of complications in children and adolescents after chest tube placement is not well documented. However, past studies in adult trauma patients undergoing chest tube placement by blunt dissection showed that 4 of 447 (1 %) had severe penetrating injuries, including lung, diaphragm, and abdominal perforation.11 Major complications in our study occurred at the time of catheter placement and demonstrate the importance of adhering to anatomic landmarks with adequate supervision by an experienced clinician during invasive procedures. We recommend placement of pigtail catheters in the midaxillary line at the nipple level for placement in the fourth intercostal space. Surprisingly, we found no increased risk of bleed- ing in patients who were coagulopathic at placement or for the duration of the catheter placement (eg, ECMO patients), despite a high incidence of bleeding reported in these patients with traditional chest tubes.l 2 We found minor complications frequently, with a higher incidence of dislodgment, kinking, and disconnection in the neonates compared with older children. It is our observation that the flexible nature of the catheters predisposes them to mechanical failures in comparison to the large, stiff Argyle tubes. During the study period, we found that use of anesthesia extension tubing decreased the tension on the suture site, and appeared to decrease the likelihood of accidental removal, kinking, and disconnection. Nevertheless, minor complications or malfunctions can still occur and clearly, the catheters need close monitoring with well-trained nursing and physician staff. The study limitations should be considered. This study is limited by incomplete documentation of all catheter data in the medical record, and we suspect that some minor complications were not recorded. We have no direct comparison to conventional chest tubes; however, very few are now placed outside the operating room in this institution. A direct comparison of conventional chest tubes to the pigtail catheters would be very useful for recommendations regarding empyema. Such a study would also be useful to determine patient comfort during both placement and use of the pigtail catheter compared with conventional chest tubes. Although we did not collect supportive data, it has been our observation that the pigtail catheters are more comfortable for patients than larger chest tubes. Percutaneous pigtail catheters are useful in the drainage of pleural air and fluid, particularly serous and chylous effusions. Empyema remains a difficult clinical problem and was not responsive to the placement of a pigtail catheter in this series. The catheters have potential complications, including perforation of vessels and organs as well as complications unique to the small flexible catheter. ACKNOWLEDGMENT: We would like to acknowledge the assistance of Debbie Ridling, RN, MS, CCNS, of Children's Hospital and Medical Center, for her efforts to improve our use of percutaneous catheters. REFERENCES l Bertino RE, Wesbey GE, Johnson R. Horner S)11drome occurring as a complication of chest tube placement. Radiology 1987; 99:745 2 Miller KS, Sahn SA. Chest tubes: indications, technique, management and complications. Chest 1987; 91: Iberti TJ, Stern PM. Chest tube thoracostomy. Crit Care Clin 1992; 8: Clinical Investigations

6 4 Fuhrman BP, Landrum BG, Ferrara TB, et al. Pleural drainage using modified pigtail catheters. C1it Care Med 1986; 14: Lawless S, Orr R, Killian A, et a!. New pigtail catheter for pleural drainage in pediatric patients. Crit Care Med 1989; 17: Ramnath RR, Heller RM, Ben-Ami T, eta!. Implications of early sonographic evaluation of parapneumonic effusions in children with pneumonia. Pediatrics 1998; 101: Alkrinawi S, Chernick V. Pleural fluid in hospitalized pediatric patients. Clin Pediatr 1996; 35:5-9 8 Allen R, Jung A, Lester P. Effectiveness of chest tube evacuation of pneumothorax in neonates. J Pediatr 1981; 99:629 9 Collop NA, Kim S, Sahn SA. Analysis of tube thoracostomy performed by pulmonologists at a teaching hospital. Chest 1997; 112: Moessinger A, Driscoll J, Wigger J. High incidence of lung perforation by chest tube in neonatal pneumothorax. J Pediatr 1978; 92: ll Millikan JS, Moore EE, Steiner E, et a!. Complications of tube thoracostomy for acute trauma. Am J Surg 1980; 140: Kanto WP. A decade of experience with neonatal extracorporeal membrane oxygenation. J Pediatr 1994; 124: CHEST I 114 I 4 I OCTOBER,